Many have had high expectations for the impact of computer-based technology on educational practice. By and large, these expectations have not been realised. It has become evident that innovative technology alone does not necessarily guarantee progress- nor perhaps even significant change- in educational practice. This has led educational researchers to place greater emphasis on cultural issues that could account for the unexpectedly limited influence of technology-enhanced learning. This perception of the relationship between technology and learning is elaborated in the first section of the paper. It is complemented by a review of an alternative conception of computing, rooted in a methodology for modelling with dependency directed at the development of construals rather than programs, that is far better aligned to the demands of developing environments for learning. The paper concludes with a discussion of the potential implications of this approach.

Abstract. This paper reports on aspects of the Playground project in which young children (age 6 to 8) are writing and sharing their own computer videogames. We discuss how structures in the kernel language influenced the design of one of the project’s playgrounds and in turn children’s thinking and use of rules. One feature of the paper is the range of children’s responses to the task of translating their ideas for games into formal rules; kernel features, such as object orientation and the use of events, at times support and at other times constrain those responses. 1

All learning environments are designed based on a set of epistemological assumptions about what knowledge should be learned. For example, most mathematics classrooms are designed to teach a certain kind of mathematical knowledge that comprises procedures that solve isolated problems quickly, and this implicitly devalues the importance of structural understanding or, put another way, of developing an appreciation of underlying mathematical models (see Lehrer & Schauble, this volume). This means that students all too often do not appreciate the need for consistency or rigor, so do not notice conflicts, and therefore cannot learn from it. Based on our research in a variety of workplace situations, we are convinced that a crucial element of knowledge required by most, if not all, people, is precisely this appreciation of underlying models. A version of mathematics that emphasizes structures has also the potential to help students understand the computational systems that are increasingly critical in today’s society, because computer systems are mathematical models—computer software is built out of variables and relationships. As technology becomes more and more advanced, and the underlying models become more and more obscure and invisible, it

environment applied to a psychopedagogic evaluation process